Predictive Maintenance for Multistage Roll-to-Roll Manufacturing Systems

The goal of this project is to develop a playbook for designing and implementing artificial intelligence-based predictive maintenance solutions, focusing on multistage roll-to-roll (R2R) manufacturing while also documenting work done in discrete manufacturing.

Problem

Many enterprises utilize schedule-based maintenance programs — which rely on statistical analysis of historical data, such as mean time between failures — to decide when equipment should be serviced or run to failure. However, with advances in sensor technologies and predictive software solutions, manufacturers want a more proactive approach. Equipment degradation has a huge impact on product quality and manufacturing continuity.

Proposed Solution

As it explores predictive maintenance for R2R manufacturing, the team will: - Demonstrate an AI-based predictive maintenance solution for R2R manufacturing by using a pilot manufacturing facility. - Deliver a playbook that details all the key steps in the process; lists best practices for manufacturers to follow when choosing a similar predictive maintenance technology stack; points out pitfalls to avoid; and highlights the business value and cost savings predictive maintenance solutions can provide.

Impact

This playbook should lead to increased adoption of predictive maintenance technology across discrete and R2R manufacturing application domains, ultimately driving improvements in manufacturing productivity and efficiency.